Have you taken a look at Richard Trudeau's The non-Euclidean revolution (with an introduction by H. S. M. Coxeter. Birkhäuser Boston, Inc., Boston, MA, 1987. xiv+269 pp. ISBN: 0-8176-3311-1)?
The first two paragraphs of the review of this book, which K. Strubecker contributed to MathSciNet, read thus:
Starting from a very detailed, critical overview of plane geometry as
axiomatically based by Euclid in his Elements, the author, in this
remarkable book, describes in an incomparable way the fascinating path
taken by the geometry of the plane in its historical evolution from
antiquity up to the discovery of non-Euclidean geometry. This
discovery, characterized by the names of Gauss, Bolyai and
Lobachevskiĭ, signified a revolution not just for geometry; the
philosophical views of space—as shaped predominantly by Kant and
generally accepted—and also epistemology underwent an unforeseen
upheaval. This "non-Euclidean revolution'', in all its aspects, is
described very strikingly here.
The book begins with a detailed critical discussion of Euclidean
explanations of concepts, definitions and axioms. The fifth axiom, in
Euclid's version not very transparent, then, in the intuitive
formulation named after John Playfair, appeared very much clearer and
convincing. The many futile attempts at proving Euclid's "parallel
axiom'' and the final solution of the parallel problem through the
discovery of non-Euclidean geometry are described in detail. Finally,
the relative freedom from contradiction in non-Euclidean geometry is
proved. Much attention is paid to thus newly acquired horizons of
knowledge and their influence on problems in natural philosophy and
particularly in physics...
Seems to me that this book might be right up your alley!